Retinal vascular junction detection and classification via deep neural networks.

Journal: Computer methods and programs in biomedicine
PMID:

Abstract

BACKGROUND AND OBJECTIVES: The retinal fundus contains intricate vascular trees, some of which are mutually intersected and overlapped. The intersection and overlapping of retinal vessels represent vascular junctions (i.e. bifurcation and crossover) in 2D retinal images. These junctions are important for analyzing vascular diseases and tracking the morphology of vessels. In this paper, we propose a two-stage pipeline to detect and classify the junction points.

Authors

  • He Zhao
    School of Environment and Natural Resource, Renmin University of China, Beijing 100872, China. Electronic address: zhaohe628@163.com.
  • Yun Sun
    School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China.
  • Huiqi Li
    School of Public Health and Community Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.